Fixed bug in derived params when OMDOT uncertainty is 0

[dlakaplan]

Closes #1688

Tried to compute Mtot from OMDOT including uncertainty in fitter.derived_params(). But in this case OMDOT had no uncertainty since it was a funcParameter. Now has explicit protection against 0 uncertainties (whether from this or a frozen model).

Related question: should the Mtot computation there happen at all if OMDOT < 0?

[codecov]

Codecov Report

Attention: 1 lines in your changes are missing coverage. Please review.

Comparison is base (c406161) 68.41% compared to head (7896a55) 68.41%.

❗ Current head 7896a55 differs from pull request most recent head 28069fe. Consider uploading reports for the commit 28069fe to get more accurate results

Files	Patch %	Lines
src/pint/fitter.py	50.00%	1 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #1689   +/-   ##
=======================================
  Coverage   68.41%   68.41%           
=======================================
  Files         104      104           
  Lines       24339    24339           
  Branches     4346     4346           
=======================================
  Hits        16651    16651           
  Misses       6600     6600           
  Partials     1088     1088           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.